3

I am really not happy with the way questions are marked as duplicate lately. A few recent examples:

  • This question could probably be closed for several reasons, but not for being a duplicate of the indicated post; OP is only giving an example, which now is happily upgraded to the main subject of the post. In other words: any example he would have given would have been taken as main subject and would have lead to marking the question as duplicate.
  • In this question, closed as duplicate of this one, the poster asks how to rename a collection of files, with a number inside the name, not a number as name (apart from its extension).

    Why isn't this a duplicate? When two different questions deliver one or more answers which are not exchangeable, and could not easily be made exchangeable by an average user, how can it be a duplicate? Appearantly the questions differ enough to produce different (correct) answers.

    The question is marked as duplicate because one of the answers works on both questions. That is: in most cases, but not all. That is turning things upside down; defining a question by one of its answers.

    When it comes to scripting/command line, with a bit of work we could create an answer that works on maybe ten different questions. When we do, would we close the nine latest questions as a duplicate and ignore existing answers on them?

  • Another example: this question could very well be a duplicate of something, I didn' t check thouroughly, but not of this one, which is specifically about editing ~/.bashrc, while the question is definitely more in general (see the other answer).

I can understand the problem of too many questions on (almost) the same topic. In an ideal situations, someone who asks a question would generalize his problem for a broader purpose, inviting to answer for a broader range of situations. However, most people do not.

Forcing questions into one original in cases where possible (correct) answers are not exchangeable is not a solution in my opinion.

11
  • If it counts for anything, I had flagged the first question as being too broad, and I still feel that's the reason why it should be closed.
    – muru
    Jul 29, 2014 at 8:04
  • @muru I totally agree, that is what I meant with "could be closed for several reasons". Jul 29, 2014 at 8:26
  • indeed the fact that i can provide code that can work on two cases don't make them equal. But that i can do so with a tool meant to be used for the overall purpose of the cases without undue complication makes them very closely related. Indeed, given the motto of Perl, I can think of more than one way to do it, but choose one that involved only one rename command.
    – muru
    Jul 29, 2014 at 18:49
  • @muru, absolutely, but if OP for whatever reason prefers not to use Perl, two different questions/answers remain ;) Jul 29, 2014 at 18:54
  • and then I must say that the OP ought to provide that detail.
    – muru
    Jul 29, 2014 at 18:56
  • @muru I don't think so, OP does not know what answers will possibly show up. Jul 29, 2014 at 19:00
  • then the argument that OP does not want to use Perl is invalid. Only one of those statements can hold at a time.
    – muru
    Jul 29, 2014 at 19:02
  • @muru on the contrary; prefer not to use is not the same as exclude. Jul 29, 2014 at 19:04
  • ah, but his preference is known to him. We can't read his mind. If he doesn't give a preference because he doesn't know what may come up, then I believe we are perfectly correct to redirect him to an existing tool that can solve his problem. That he refuses to acknowledge that said tool can solve his problem is not my or your or anybody else's problem, but solely his own.
    – muru
    Jul 29, 2014 at 19:07
  • @muru well..., not if the question is not the same in the first place, which is my point. In his comment he makes clear he looked at the other question by the way (We are still discussing question 2 right :)) Jul 29, 2014 at 19:08
  • that is the whole point of general purpose tools like rename. That they be malleable enough to handle a variety of cases. Not acknowledging that is reducing a modern computer with all the accompanying high level software to counting with fingers.
    – muru
    Jul 29, 2014 at 19:11

1 Answer 1

1

The second question was closed by me, against a question I had answered.

There can be answers that work on both questions, but not necessarily and in general it would take another approach.

Both questions are exactly the same. We are taking part of a filename (the original still has an extension) and we are padding the number with zeros to circumvent natural-number sorting issues.

Adding a suffix of track_ is an extension of the problem but it makes no practical difference that isn't already handled by how we're dealing with the extension from the original. To repeat myself, any required technical alteration to deal with a new prefix can be observed from how we're dealing with the extension.

And that's why I still consider this an exact duplicate.

If you're still not with me, what would you do if somebody asked the same question but with tracks_##.mp3 instead of track_##.mp3? I don't think we need to cater to complete idiots.

(And to the ~"Oh, sure you know it's a duplicate"): that really makes no difference. Closing something as a duplicate isn't telling the OP they're wrong or stupid, just that their question has been answered elsewhere in the past. In this case it's me who has technically checked to make sure the master does answer the duplicate question. That's what I did here.

5
  • 1
    I won't deny the questions are related to each other, but if that would be the standard, the number of questions on AU could be reduced to (I guess) a couple of hundreds. In many cases, answers of questions can easily be derived from each other with enough knowledge. That includes for example the many questions on (the relation between) scripts, paths, language extensions and shebangs. Jul 29, 2014 at 8:27
  • 1
    Furthermore, reacting on your latest comment below the question ("The question is exactly the same, the input data is slightly different..."): This is exactly my point: from your point of view, the difference between the questions is the same as the difference between cd to different directories. I can assure you that is a point of view of someone of your level, which is quite different from an average user. I have no doubt you can produce an answer is a possible solution to a much broader range than these two questions. That does not make the questions the same, but proves your quality. Jul 29, 2014 at 18:58
  • 1
    I've edited but I think you're hugely overstating the technical differences here. To show that, look at any of the answers from the closed question, remove track_ from them and they all magically Just Work™ on the master question. Seriously, this isn't rocket science.
    – Oli Mod
    Jul 30, 2014 at 14:49
  • 1
    I already noticed your answer works on both examples, but marking it as duplicate because of that is tricky; not all answers do (mine doesn't for example). Furthermore: I remember in the days I used Windows (or it might have been Mac) I owned a scanner, of which the software saved images, named after the software + (main) version number + underscore + file number + file extension. The first question has no working answer to that, as far as I can see, while the second one does, as a result of the fact that the question specifically asks how to deal with prefixes. (to be continued...) Jul 31, 2014 at 8:43
  • ...Of course that can be easily be solved, but solving differences between questions is not the idea of a duplicate in my opinion, and although it is straightforward in principle, assuming that the average user will be able to do that is a bit optimistic I guess. Jul 31, 2014 at 8:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .